Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@lighthouse-mcp list my lighthouse instances and their status"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Server for Tencent Cloud Lighthouse（轻量应用服务器）管理。
让 AI Agent（Claude Code / OpenClaw / Cursor）通过自然语言管理你的 Lighthouse 服务器。
工具 | 说明 |
| 查询所有可用地域 |
| 查询实例列表 |
| 获取监控数据（CPU/内存/带宽/磁盘IO） |
| 查看防火墙规则 |
| 添加防火墙规则 |
| 删除防火墙规则 |
| 查看快照列表 |
| 远程执行命令（TAT） |
| 查询命令执行状态 |
Related MCP server: hcloud-mcp
npm install -g lighthouse-mcp
# 或
npx lighthouse-mcp设置腾讯云 API 密钥：
export TENCENT_SECRET_ID="your_secret_id"
export TENCENT_SECRET_KEY="your_secret_key"或创建 .env 文件：
TENCENT_SECRET_ID=your_secret_id
TENCENT_SECRET_KEY=your_secret_key密钥获取：腾讯云控制台 > API密钥管理
添加到 ~/.claude/claude_desktop_config.json 或项目 .mcp.json：
{
"mcpServers": {
"lighthouse": {
"command": "npx",
"args": ["-y", "lighthouse-mcp"],
"env": {
"TENCENT_SECRET_ID": "your_secret_id",
"TENCENT_SECRET_KEY": "your_secret_key"
}
}
}
}添加到 openclaw.json 的 tools.mcp 配置：
{
"tools": {
"mcp": {
"lighthouse": {
"command": "node",
"args": ["dist/index.js"],
"cwd": "/path/to/lighthouse-mcp",
"env": {
"TENCENT_SECRET_ID": "...",
"TENCENT_SECRET_KEY": "..."
}
}
}
}
}你: 我有几台服务器？状态怎么样？
Agent: [调用 describe_instances] → 你有 2 台服务器，上海那台运行中...
你: 帮我把上海的服务器防火墙打开 443 端口
Agent: [调用 create_firewall_rules] → 已添加 TCP 443 规则
你: 最近的 CPU 使用率怎么样？
Agent: [调用 get_monitor_data] → 最近 6 小时 CPU 均值 12%...@modelcontextprotocol/sdk — MCP 协议 SDK
tencentcloud-sdk-nodejs-lighthouse — 腾讯云 Lighthouse SDK
TypeScript + ES Modules
MIT
为实例添加防火墙规则
Inputs: region (required), instanceId (required), firewallRules (required).
删除实例的防火墙规则
Inputs: region (required), instanceId (required), firewallRules (required).
查询 TAT 命令执行任务状态
Inputs: region (required), invocationTaskId (required).
查询实例的防火墙规则
Inputs: region (required), instanceId (required).
查询轻量应用服务器实例列表
Inputs: region (required), instanceIds, limit, offset.
查询所有可用的 Lighthouse 地域列表
Inputs: no named parameters.
查询实例的快照列表
Inputs: region (required), instanceId (required).
在实例上远程执行命令（通过 TAT）
Inputs: region (required), instanceId (required), command (required), systemType.
获取实例监控数据（CPU/内存/带宽/磁盘IO等）
Inputs: region (required), instanceId (required), indicators (required), startTime, +1 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 9 tools
Each tool targets a distinct operation: firewall management (create/delete/describe), command execution and task status, instance listing, region listing, snapshots, and monitoring. No two tools overlap in purpose.
All tools follow a consistent verb_noun pattern in snake_case (e.g., create_firewall_rules, describe_instances, get_monitor_data). Verbs like describe, create, delete, execute, get are used predictably.
With 9 tools, the server is well-scoped for managing key aspects of Lighthouse instances: firewall, command execution, monitoring, and listing resources. The number is appropriate without being overwhelming.
The tool set covers firewall, command execution, monitoring, and listing of instances/snapshots/regions, but lacks instance lifecycle operations (create, delete, stop, start) and snapshot CRUD (only describe). This leaves notable gaps for full instance management.
Secure and manage MCP servers through the Speakeasy AI Control Plane from your agent.
MCP server for AI dialogue using various LLM models via AceDataCloud
MCP server for ByteDance Seedream AI image generation
MCP server for Appcircle mobile CI/CD platform.
